WO2025101504A1

WO2025101504A1 - Machine learning-based model for online stimulation efficiency monitoring

Info

Publication number: WO2025101504A1
Application number: PCT/US2024/054547
Authority: WO
Inventors: Roman Vladimirovich Korkin; Abdul Muqtadir KHAN
Original assignee: Schlumberger Canada Ltd; Services Petroliers Schlumberger SA; Schlumberger Technology BV; Schlumberger Technology Corp
Current assignee: Schlumberger Canada Ltd; Services Petroliers Schlumberger SA; Schlumberger Technology BV; Schlumberger Technology Corp
Priority date: 2023-11-08
Filing date: 2024-11-05
Publication date: 2025-05-15
Anticipated expiration: 2026-05-08

Abstract

Embodiments presented provide for using machine learning-based models in hydrocarbon recovery projects. In embodiments, stimulation efficiency is determined through a machine learning based model to enhance hydrocarbon recovery in wellbore applications. A method may include providing a computer model configured to use artificial intelligence. The method may include feeding the computer model data obtained from one of an optic-acquired vibration dataset and a fiber optic cable dataset, and training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset. The method may include obtaining surface data and inputting the surface data into the computer model, and predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

Description

MACHINE LEARNING-BASED MODEL FOR ONLINE STIMULATION EFFICIENCY MONITORING

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application

No. 63/597111 , filed on November 8, 2023.

FIELD OF THE DISCLOSURE

[0002] Aspects of the disclosure relate to machine learning models. More specifically, aspects of the disclosure relate to machine learning models that determine stimulation efficiency in hydrocarbon recovery projects.

BACKGROUND

[0003] Hydraulic fracturing operation requires careful monitoring of multiple parameters to perform stimulation jobs efficiently. In order to perform an efficient job, a uniform stimulation of all intervals during a stage is required. To achieve this, pumping may be performed with an amount of proppant added to water. In embodiments, diversion of fluids with chemical diverters may be performed. If a single stage is designed to treat a long interval, then chemical diverters may be pumped to ensure its uniform stimulation, i.e., all clusters and perforated holes take approximately equal (or required) fracturing fluid and proppant, at the end of the fracturing treatment. Generally, uniform slurry distribution over perforation clusters leads to higher production rates. There are multiple observational parameters that effect stimulation efficiency. One example, called the Nolte Smith log method, allows predicting screenout conditions by observing a typical slope of the tangent on the log (Net Pressure)-vs.-log (Time) plot.

[0004] In embodiments, the working principle of chemical diverters is to seat the chemical diverters at the fractures/perforation holes taking fluid to generate wellbore skin at these points. This skin is associated with wellbore pressure increases. Efficiency of the diverters may be measured by the pressure that arises after pumping. Such pressure variation; however, is not always a reliable indication due to the interplay of multiple, near wellbore, dynamics affecting the pressure’s behavior. An alternative method of measuring efficiency is to provide a high frequency pressure monitoring system.

[0005] While stimulation efficiency is an important value to be able to calculate, conventional analysis has difficulty in providing a consistent value. There is a need to provide a method and apparatus to be able to model stimulation efficiency.

[0006] There is a need to provide an apparatus and methods that are easier to operate than conventional apparatus and methods and do not have the requirements of installing an extensive high pressure monitoring system as required by conventional systems.

[0007] There is a further need to provide apparatus and methods that do not have the drawbacks discussed above including using extensive chemical diverters.

[0008] There is a still further need to reduce economic costs associated with operations and apparatus described above with conventional tools and overall economic costs associated with calculating stimulation efficiency.

SUMMARY

[0009] So that the manner in which the above recited features of the present disclosure can be understood in detail, a more particular description of the disclosure, briefly summarized below, may be had by reference to embodiments, some of which are illustrated in the drawings. It is to be noted that the drawings illustrate only typical embodiments of this disclosure and are therefore not to be considered limiting of its scope, for the disclosure may admit to other equally effective embodiments without specific recitation. Accordingly, the following summary provides just a few aspects of the description and should not be used to limit the described embodiments to a single concept. [0010] In one example embodiment of the disclosure a method is provided. The method may comprise providing a computer model configured to use artificial intelligence. The method may further comprise feeding the computer model data obtained from one of a optic-acquired vibration dataset and a fiber optic cable dataset. The method may further comprise training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset. The method may further comprise obtaining surface data and inputting the surface data into the computer model. The method may further comprise predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

[0011] In another example embodiment of the disclosure, an article of manufacture is provided, having a non-volatile memory, a set of instructions encoded onto the memory that may be read and performed by a computer, the set of instructions comprising a method. The method recited by the article of manufacture may comprise providing a computer model configured to use artificial intelligence. The method recited by the article of manufacture further configured to recite feeding the computer model data obtained from one of a optic-acquired vibration dataset and a fiber optic cable dataset. The method recited by the article of manufacture further configured to recite training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset. The method recited by the article of manufacture further configured to recite obtaining surface data an inputting the surface data into the computer model. The method recited by the article of manufacture further configured to recite predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

[0012] In another example embodiment, a method is disclosed. The method may comprise providing a computer model configured to use artificial intelligence. The method may further comprise feeding the computer model data obtained from one of a optic- acquired vibration dataset and a fiber optic cable dataset. The method may further comprise training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset. The method may further comprise obtaining surface data on a proppant, a proppant supply rate, and a proppant concentration provided to the wellbore, and inputting the surface data into the computer model. The method may further comprise predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

BRIEF DESCRIPTION OF THE DRAWINGS

[0013] So that the manner in which the above recited features of the present disclosure can be understood in detail, a more particular description of the disclosure, briefly summarized above, may be had by reference to embodiments, some of which are illustrated in the drawings. It is to be noted; however, that the appended drawings illustrate only typical embodiments of this disclosure and are therefore not be considered limiting of its scope, for the disclosure may admit to other equally effective embodiments.

[0014] FIG. 1 is a non-dimensional slurry distribution for distribution of six clusters per stage.

[0015] FIG. 2 is a non-dimensional slurry distribution over time in seconds.

[0016] FIG. 3 is a non-dimensional slurry distribution diverter plot.

[0017] FIG. 4 is a non-dimensional slurry distribution versus perforation cluster data in acid fracturing.

[0018] FIG. 5 is a method for calculating stimulation efficiency through a machine learning based model in one example embodiment of the disclosure. [0019] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures (“FIGS”). It is contemplated that elements disclosed in one embodiment may be beneficially utilized on other embodiments without specific recitation.

DETAILED DESCRIPTION

[0020] In the following, reference is made to embodiments of the disclosure. It should be understood; however, that the disclosure is not limited to specific described embodiments. Instead, any combination of the following features and elements, whether related to different embodiments or not, is contemplated to implement and practice the disclosure. Furthermore, although embodiments of the disclosure may achieve advantages over other possible solutions and/or over the prior art, whether or not a particular advantage is achieved by a given embodiment is not limiting of the disclosure. Thus, the following aspects, features, embodiments and advantages are merely illustrative and are not considered elements or limitations of the claims except where explicitly recited in a claim. Likewise, reference to “the disclosure” shall not be construed as a generalization of inventive subject matter disclosed herein and should not be considered to be an element or limitation of the claims except where explicitly recited in a claim.

[0021] Although the terms first, second, third, etc., may be used herein to describe various elements, components, regions, layers and/or sections, these elements, components, regions, layers and/or sections should not be limited by these terms. These terms may be only used to distinguish one element, components, region, layer or section from another region, layer or section. Terms such as “first”, “second” and other numerical terms, when used herein, do not imply a sequence or order unless clearly indicated by the context. Thus, a first element, component, region, layer or section discussed herein could be termed a second element, component, region, layer or section without departing from the teachings of the example embodiments.

[0022] When an element or layer is referred to as being “on,” “engaged to,” “connected to,” or “coupled to” another element or layer, it may be directly on, engaged, connected, coupled to the other element or layer, or interleaving elements or layers may be present. In contrast, when an element is referred to as being “directly on,” “directly engaged to,” “directly connected to,” or “directly coupled to” another element or layer, there may be no interleaving elements or layers present. Other words used to describe the relationship between elements should be interpreted in a like fashion. As used herein, the term “and/or” includes any and all combinations of one or more of the associated listed terms and should not be considered limiting.

[0023] Some embodiments will now be described with reference to the figures. Like elements in the various figures will be referenced with like numbers for consistency. In the following description, numerous details are set forth to provide an understanding of various embodiments and/or features. It will be understood; however, by those skilled in the art, that some embodiments may be practiced without many of these details, and that numerous variations or modifications from the described embodiments are possible. As used herein, the terms “above” and “below”, “up” and “down”, “upper” and “lower”, “upwardly” and “downwardly”, and other like terms indicating relative positions above or below a given point are used in this description to more clearly describe certain embodiments.

[0024] Stimulation efficiency is an important value in hydrocarbon recovery projects. The parameter, characterizing stimulation efficiency is a non-dimensional slurry distribution (NSD), determined as:

where n is the number of clusters, q_t are the individual clusters flowrates (normalized onto a total flowrate). NSD value varies from 1 (fully uniform distribution) to zero (only one cluster is stimulated). The different distributions and the respective NSD values are shown in FIG. 1. Though this parameter does not assume exact prediction of the flow rate distribution, it reflects the overall uniformity quite well. It is shown that the same NSD may correspond to a slightly different distribution, though the overall uniformity looks similar. As will be understood, other formulations for NSD values are considered as part of the disclosure, and the equation presented above should not be considered limiting.

[0025] In aspects of the disclosure, a machine learning-based model may be used to determine values, such as stimulation efficiency. Such calculations will provide for greater accuracy in prediction of hydrocarbon recovery. In aspects of the disclosure, the training is being performed with the recurrent neural network (though may be performed with a classical neural network if the features are properly engineered, or with the use of transformers, or other architecture). Further, in one or more embodiments, the computer model is trained based on transfer learning techniques using field data and synthetic data based on physics. Values used in embodiments include pressure, rate, concentration, total amount of proppant, clean fluid, and slurry pumped till the current moment of time, and wellbore data (internal diameter, average over stage measured depth and true vertical depth). These parameters are properly prepared to represent physically non- dimensional features and the network trained to predict NSD, matching with the NSD received from the fiberoptic data, which is the basis for the machine learning approach. Real-time NSD monitoring may indicate whether a large portion of clusters is being stimulated. This may be used as an indicator to pump a chemical diverter (if the efficiency is low), or to continue pumping if the uniformity reaches the required cluster efficiency. Another application is to use the NSD evolution for a given treatment stage and implement observations/lessons for the subsequent stage. For instance, aspects of the disclosure may allow for stimulation design alteration, while NSD dynamics, before and after diverter pumping, may indicate whether the diverter performed as planned (some of intervals/clusters plugged and the NSD drops), etc. As will be understood, in some aspects of the disclosure, diverter pumping may occur. In other embodiments, diverter pumping may not occur.

[0026] Aspects of the disclosure suggest that one unique parameter allows for the performance of the overall evaluation of a simulation efficiency. In embodiments, a neural network based model is provided. The model is trained on data, such as fiber optic- acquired vibration data, received during fracturing from a well-deployed (permanently or temporarily) fiber optic sensor and surface data (pressure, rate, concentration). In an inference mode of the model, the model allows for stimulation efficiency determination using only wellbore completion data, pressure, rate, and concentration data. The output of the model is a one-dimensional parameter, varying from 1 (uniform slurry distribution over all clusters along the stage) down to 0 (only one cluster is being stimulated).

[0027] Aspects of the disclosure may use different types of networks. For example, one embodiment may use convolutional networks (CNN). In another embodiment, a recurrent network may be used. Aspects of the disclosure may provide for different types of analysis. In one example embodiment, wellbore completion data may be used, i.e. measured and true vertical depths, and the wellbore diameter. In another example embodiment, pumping data is not used as they are, but their nondimensional physical variables are built. Thus, the results are applicable to a wider range of wells, with different depths, trajectories, diameters, and pressures. Using the time-series training, based on the recurrent neural network (transformer-based architecture can also be used) calculations using different values were checked. In some embodiments, recurrent neural networks operate better than the convolutional neural networks. In some embodiments, the current algorithm does not stack the data during long periods, so that it performs much faster and does not require any special computing device or devices, neither in the inference mode nor in training modes. In one or more embodiments, analysis by the computer model may include at least one of supervised learning, unsupervised learning, deep learning, transfer learning, knowledge graph, regression analysis, clustering techniques, bagging techniques or boosting techniques.

[0028] Aspects of the disclosure have been successfully trained, validated, and tested on separate datasets. The typical results are shown in FIG. 2. Here, the NSD varies in the range 0.7-0.95 most of the time for job 1 , and shows lower values for job 2. Reasonable outcomes of the predicted NSD versus those computed from the cluster distributions obtained from the fiber optic-acquired data are possible. Despite the original model having been trained on stages with different numbers of clusters, the model in the inference mode does not use the number of clusters in the stage as a parameter. As a result, it is applicable for a wider range of completions than multi-stage, perforation-based completions. In the case of open holes, it predicts uniformity over the whole stage rather than over clusters.

[0029] In FIG. 3, a project with a diverter plot is shown. In this given example, the pressure rise has been observed just after diverter pumping, which may indicate (though not reliably as stated above) that the diversion operation was successful. The successful diversion only means that few intervals from the original number were plugged, but cannot indicate whether the new intervals are opened and successfully stimulated. Also, chemical diverters frequently fail right after their placement due to unseating or degradation effects. Using only pressure increase to evaluate diversion efficiency can be misleading. The NSD, contrary to the pressure, shows that the stimulation efficiency can drop during the first stage of pumping from 0.8 down to 0.72. Then, during diverter pumping, drops even lower to 0.68, which indicated good diverter performance, in terms of plugging efficiency of current fractures, in agreement with the pressure analysis. Then, NSD starts increasing, though not reaching the original level, meaning that some intervals do not accept fluid anymore and the slurry is spread among the remaining depths.

[0030] The general trend for NSD to decrease during long term pumping is common; as some intervals which accepted more slurry are prone to the erosion; thus, should accept more and more fluid compared to the intervals which have not been eroded.

[0031] The diverter performance can be estimated quantitatively using NSD as below: i = + Bi

Ai = NSD₁ - NSD₂

Bi = NSD₃ - NSD₂, where, ]I is diverter efficiency, in terms of NSD and cluster stimulation

A_t is fractures plugged in the ith diverter pill, in terms of NSD

B_t is new fractures opened after the ith diverter pill, in terms of NSD

NSD is the representative NSD value before pumping the ith diverter pill

NSD₂ is the NSD value right after the ith diverter pill is placed at perforations

NSD₃ is the stabilized NSD value while pumping the (i+1 )th pumping cycle and for n number of diverter pills, the total diverter efficiency can be calculated as £ ₌₀ rji.

[0032] The model can also be applied for acid fracturing projects, where no perforation clusters are used (open hole), and in vertical wells with long, perforated intervals, in which proppant is not pumped. Such an example is shown in FIG. 4. Retarded acid was used and should have worked only with some delay. The NSD allows the following observation:/shows that it rises from 0.67 to more than 0.8, with some delay. Further, the NSD shows some sagging during lower rates which are expected for the efficiency, as well as a decreasing trend.

[0033] Referring to FIG. 5, a method 500 for calculating stimulation efficiency. The method may include, at 502, providing a computer model configured to use artificial intelligence. The method may further comprise, at 504, feeding the computer model data obtained from one of a optic-acquired vibration dataset and a fiber optic cable dataset. The method may further comprise, at 506, training the computer model on the optic- acquired vibration dataset and the fiber optic cable dataset. The method may further comprise, at 508, obtaining surface data and inputting the surface data into the computer model. The method may further comprise, at 510, predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

[0034] Aspects of the claims are disclosed. The recitation of the features of the claims should not be considered to limit the disclosure. In one example embodiment of the disclosure a method is provided. The method may comprise providing a computer model configured to use artificial intelligence. The method may further comprise feeding the computer model data obtained from one of a optic-acquired vibration dataset and a fiber optic cable dataset. The method may further comprise training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset. The method may further comprise obtaining surface data and inputting the surface data into the computer model. The method may further comprise predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

[0035] In another example embodiment, the method may be performed wherein the surface data contains at least one of data on a proppant, a proppant supply rate, and a proppant concentration provided to the wellbore. [0036] In another example embodiment, the method may be performed wherein the computer model is one of a neural network, a predictive modeling architecture, and a machine learning architecture.

[0037] In another example embodiment, the method may further comprise optimizing the wellbore treatment based upon the surface data.

[0038] In another example embodiment, the method may further comprise a diversion pumping for the wellbore based upon calculations from the computer model.

[0039] In another example embodiment, the method may be performed wherein analysis by the computer model includes at least one of supervised learning, unsupervised learning, deep learning, transfer learning, knowledge graph, regression analysis, clustering techniques, bagging techniques, and boosting techniques.

[0040] In another example embodiment, the method may be performed wherein the computer model is trained on synthetic data.

[0041] In another example embodiment, the method may be performed wherein the computer model is trained on field data from a previous project.

[0042] In another example embodiment, the computer model is trained based on transfer learning techniques using field data and synthetic data based on physics. [0043] In another example embodiment, the method may be performed wherein the computer model is one based upon a web architecture, a server architecture, and a personal computer architecture.

[0044] In another example embodiment, the method may be performed wherein the wellbore treatment includes at least one of proppant fracturing, acid fracturing, matrix acidizing, sand control, or water control treatments.

[0045] In another example embodiment, the method may be performed wherein the wellbore has at least one section that is vertically oriented, horizontally oriented, and inclined at an angle.

[0046] In another example embodiment, the method may be performed wherein the wellbore is one of a cemented cased hole, an open hole, an open hole with fracturing sleeves, a wellbore with isolation packers, and has pre-perforated liners.

[0047] In another example embodiment of the disclosure, an article of manufacture is provided, having a non-volatile memory, a set of instructions encoded onto the memory that may be read and performed by a computer, the set of instructions comprising a method. The method recited by the article of manufacture may comprise providing a computer model configured to use artificial intelligence. The method recited by the article of manufacture further configured to recite feeding the computer model data obtained from one of a optic-acquired vibration dataset and a fiber optic cable dataset. The method recited by the article of manufacture further configured to recite training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset. The method recited by the article of manufacture further configured to recite obtaining surface data and inputting the surface data into the computer model. The method recited by the article of manufacture further configured to recite predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

[0048] In another example embodiment, the method may be performed wherein the computer model is one of a neural network, a predictive modeling architecture, and a machine learning architecture.

[0049] In another example embodiment, the method may be performed wherein the analysis by the computer model includes at least one of supervised learning, unsupervised learning, deep learning, transfer learning, knowledge graph, regression analysis, clustering techniques, bagging techniques, and boosting techniques.

[0050] In another example embodiment, a method is disclosed. The method may comprise providing a computer model configured to use artificial intelligence. The method may further comprise feeding the computer model data obtained from one of a optic- acquired vibration dataset and a fiber optic cable dataset. The method may further comprise training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset. The method may further comprise obtaining surface data on a proppant, a proppant supply rate and a proppant concentration provided to the wellbore and inputting the surface data into the computer model. The method may further comprise predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

[0051] In another example embodiment, the method may be performed wherein analysis by the computer model includes at least one of supervised learning, unsupervised learning, deep learning, transfer learning, knowledge graph, regression analysis, clustering techniques, bagging techniques, and boosting techniques.

[0052] In another example embodiment, the method may be performed wherein the wellbore is one of a cemented cased hole, an open hole, an open hole with fracturing sleeves, a wellbore with isolation packers and has pre-perforated liners. [0053] The foregoing description of the embodiments has been provided for purposes of illustration and description. It is not intended to be exhaustive or to limit the disclosure. Individual elements or features of a particular embodiment are generally not limited to that particular embodiment, but, where applicable, are interchangeable and can be used in a selected embodiment, even if not specifically shown or described. The same may be varied in many ways. Such variations are not to be regarded as a departure from the disclosure, and all such modifications are intended to be included within the scope of the disclosure.

[0054] While embodiments have been described herein, those skilled in the art, having benefit of this disclosure, will appreciate that other embodiments are envisioned that do not depart from the inventive scope. Accordingly, the scope of the present claims or any subsequent claims shall not be unduly limited by the description of the embodiments described herein.

Claims

CLAIMS What is claimed is:

1. A method, comprising: providing a computer model configured to use artificial intelligence; feeding the computer model data obtained from one of a optic-acquired vibration dataset and a fiber optic cable dataset; training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset; obtaining surface data and inputting the surface data into the computer model; and predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

2. The method according to claim 1 , wherein the surface data contains at least one of data on a proppant, a proppant supply rate, and a proppant concentration provided to the wellbore.

3. The method according to claim 1 , wherein the computer model is one of a neural network, a predictive modeling architecture, and a machine learning architecture.

4. The method according to claim 1 , further comprising optimizing the wellbore treatment based upon the surface data.

5. The method according to claim 1 , further comprising planning a diversion pumping for the wellbore based upon calculations from the computer model.

6. The method according to claim 1 , wherein analysis by the computer model includes at least one of supervised learning, unsupervised learning, deep learning, transfer learning, knowledge graph, regression analysis, clustering techniques, bagging techniques, and boosting techniques.

7. The method according to claim 1 , wherein the computer model is trained on synthetic data.

8. The method according to claim 1 , wherein the computer model is trained on field data from a previous project.

9. The method according to claim 1 , wherein the computer model is trained based on transfer learning techniques using field data and synthetic data based on physics.

10. The method according to claim 1 , wherein the computer model is one based upon a web architecture, a server architecture, and a personal computer architecture.

11 . The method according to claim 1 , wherein the wellbore treatment includes at least one of proppant fracturing, acid fracturing, matrix acidizing, sand control, or water control treatments.

12. The method according to claim 1 , wherein the wellbore has at least one section that is vertically oriented, horizontally oriented, and inclined at an angle.

13. The method according to claim 1 , wherein the wellbore is one of a cemented cased hole, an open hole, an open hole with fracturing sleeves, a wellbore with isolation packers, and has pre-perforated liners.

14. An article of manufacture, having a non-volatile memory, a set of instructions encoded onto the memory that may be read and performed by a computer, the set of instructions comprising a method, comprising: providing a computer model configured to use artificial intelligence; feeding the computer model data obtained from one of a optic-acquired vibration dataset and a fiber optic cable dataset; training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset; obtaining surface data an inputting the surface data into the computer model; and predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

15. The article of manufacture according to claim 14, wherein the computer model is one of a neural network, a predictive modeling architecture and a machine learning architecture.

16. The article of manufacture according to claim 1 , wherein analysis by the computer model includes at least one of supervised learning, unsupervised learning, deep learning, transfer learning, knowledge graph, regression analysis, clustering techniques, bagging techniques and boosting techniques.

17. A method, comprising: providing a computer model configured to use artificial intelligence; feeding the computer model data obtained from one of a optic-acquired vibration dataset and a fiber optic cable dataset; training the computer model on the optic-acquired vibration dataset and the fiber optic cable dataset; obtaining surface data on a proppant, a proppant supply rate, and a proppant concentration provided to the wellbore, and inputting the surface data into the computer model; and predicting stimulation efficiency of a wellbore treatment for the wellbore based upon the surface data.

18. The method according to claim 17, wherein analysis by the computer model includes at least one of supervised learning, unsupervised learning, deep learning, transfer learning, knowledge graph, regression analysis, clustering techniques, bagging techniques, and boosting techniques.

19. The method according to claim 17, wherein the wellbore is one of a cemented cased hole, an open hole, an open hole with fracturing sleeves, a wellbore with isolation packers, and has pre-perforated liners.

20. The method as illustrated and described.

21 . The apparatus as illustrated and described.